|For the si allele:|
|si Allele (MGI)||Gene (MGI)||All Alleles (MGI)|
Silver (si, chromosome 10) ( Plate 3-B) was reported first by Hagedoorn ( 1912) who claimed it occurred twice in his colony. According to Hagedoorn animals homozygous for this recessive determinant, which he called f, did not differ from Ff or FF animals until after they had moulted and their second coat was interspersed with white hairs. These were either distributed evenly over the entire body or were limited to patches on the back and sides. 13
Because Hagedoorn's allele was lost, it was not possible to determine whether it was the same as the one obtained from an English fancier by Dunn and Thigpen 15 years later. This si allele had been represented in the English fancy for several years and was known either as "silver grey" or "silver brown." Silver grey animals appeared to be black mice displaying a mixture of white-tipped, all-white, and black and white hairs. Silver brown mice "were of a lighter shade than the standard chocolate, and also showed white-tipped and some all-white hairs, although they were never silvered to such an extent as the silver greys" ( Dunn and Thigpen, 1930). 14
The "purebred silver greys" ( a/a;B/B;si/si) studied by Dunn and Thigpen were variable in phenotype. Indeed, dark, medium, and light silver greys were recognized by the fancy. Some young a/a;B/B;si/si mice were black in their first pelage, although a few silvered hairs could usually be seen on the head, behind the ears, and on the flanks, while the initial coat of others, usually those which became "light silvers," was well silvered. All the mice became progressively lighter (more silvered) with age, the males showing more silvering than the females.
Although silver greys usually exhibit the silvering effect throughout their coat, the belly is lighter than the back. The fur contains several kinds of hairs: all white, all black (these probably only in dark silvers), black hairs with white tips, and hairs with several white and grey or black bands. The all-white hairs have no pigment whereas in the all-black and black hairs with white tips the pigment granules are reduced in number and scattered. The banded hairs display white areas alternating with sparsely pigmented ones ( Dunn and Thigpen, 1930). Silvering thus seems to be the consequence of a reduction in the number of melanin granules, and of their complete absence from certain areas and from some hairs. Light silvers have more completely white hairs and larger unpigmented zones than darker silvers ( Dunn and Thigpen, 1930).
Hairs from silver brown ( a/a;b/b;si/si) mice display the same sorts of pigmented hairs as described above for silver greys, except that "the proportions of white hairs and of white in colored hairs are generally much less" ( Dunn and Thigpen, 1930). Nevertheless, in black mice heterozygous for brown the effect of si is greatly intensified; a/a;B/b;si/si animals are lighter than either B/B or b/b silver mice ( Dunn and Thigpen, 1930). According to Grüneberg ( 1952) the whole underfur of a/a;B/b;si/si mice is practically white and "the animals resemble 'reverse agoutis', the hair bases being light, the distal parts of the hairs dark." Moreover, a similar effect has been observed by Grüneberg in at/at;B/b;si/si animals. Aside from the fact that this heterozygous effect of b on the expression of silver represents an exception to the rule that B is always dominant to b, it is interesting because its effect on the phenotype is unexpected, i.e., since a/a;b/b;si/si mice usually display less silvering than a/a;B/B;si/si animals, one would anticipate any heterozygous effect of b likewise to be in the direction of less silvering.
Silver agoutis ( A/;B/B;si/si), like silver greys, are reported by Dunn and Thigpen to be "variable in appearance, some being well silvered, others being very similar to black agouti, except for white or light bases in the hairs which serves as a distinguishing mark of silver agouti." The yellow band on the agouti hair is not visibly influenced. Silver yellow ( Ay/;si/si) mice also display the same variation in intensity of silvering as silver blacks. Moreover, as in the case of agouti, the predominant effect of silver on yellow seems to be at the base of the fur which becomes very light, although white tips and silvered shafts likewise occur. Grüneberg ( 1952) notes that both agouti and yellow silver mice "have a more or less white underfur." Whether the B/b heterozygous condition influences the expression of silver on agouti and yellow backgrounds is not stated.
One of the most interesting observations of Dunn and Thigpen ( 1930) is that in contrast to the situation in nonagouti silver mice, where the animals became progressively more silvered, in agouti and yellow silver mice the silvering decreases markedly as the animals get older. In fact, Dunn and Thigpen report that some silver agoutis became almost indistinguishable from black agoutis. This finding is of particular importance insofar as the etiology of silvering is concerned. If silvering merely was a consequence of a loss of pigment cells, as it thought to be the case in man, one would expect that its severity would increase with age on all backgrounds. The fact that it does not and, indeed, the fact that its severity decreases with age on some backgrounds, indicates that the condition cannot be explained simply by the demise of pigment cells. Furthermore, inasmuch as the genetic background affecting this expression of silver is determined primarily by the agouti locus, a locus known to influence the follicular environment (see Chapter 2, Section I, E), it appears that whatever affect this locus has on this environment must also somehow influence the expression of si/si.
Superficially the expression of silver resembles that of light (Blt; see Chapter 3, Section I, G) not only because both mutations produce an overall dilution in pigmentation 15 but because at the end of the hair cycle all light hair bulbs and many a/a;B/B;si/si follicles (e.g., those which are not pigmented) do not possess clear cells and are indistinguishable from those in white spotted areas (Chase, personal communication). Nevertheless, these coat-color mutants behave differently. In a very simple but significant experiment Chase (see Quevedo and Chase, 1958) was able to deplete the pigment cell populations of a/a;B/B;si/si hair follicles through repeatedly inducing hair cycles by replucking immediately at the end of each cycle. Eventually completely white hairs were regrown. In contrast, attempts to deplete the melanocyte population of light mice by this means proved unsuccessful ( Quevedo and Chase, 1958). Of course one wonders what the effect would be of subjecting agouti and yellow silver mice to this plucking treatment? On the basis of Dunn and Thigpen's observation that these mice get darker with age, one might anticipate a decrease rather than an increase in silvering.
Chase and Rauch ( 1950) reported that "a moderate grade of silver resembles in appearance and animal treated with 300 r on inactive follicles." Thus, a silver animals treated with 400 r responded as if it already had a 300-r treatment. The net effect was the production of more grey than results from the treatment of a black mouse with 400 r. Since the silver mice employed in this study were also on a nonagouti black background (Chase, personal communication), it again raises the question whether Ay/;si/si and A/;si/si genotypes would respond similarly.
Clearly the si mutation has not received the attention it deserves. On the basis of the available evidence its effect appears to stem from some breakdown in the intimate relationship which normally prevails between pigment cells and the epithelial cells of the follicle; a breakdown which seems to result in the inability of melanocytes to differentiate fully. 16 A similar hypothesis was initially suggested as being responsible for the greying effects of low dosages of X-irradiation (Chase, personal communication) and could be the basis of greying in man as well.
Finally it should be noted that very little is known about the factors, genetic or otherwise, which are responsible for the variations in the intensity of silvering. Thus it remains uncertain whether modifying genes in the strict sense, and/or "minor" silvering genes influence the expression of the si/si genotype ( Grüneberg, 1952).